About a month ago I read this excellent article on the fall from grace of the name ‘Hilary’. That alerted me to the fact that you can easily get US Social Security records on frequency of first names right back to the 1880s. Although Hilary focused on the raw numbers, I commented that I’d be interested in seeing the behaviour with respect to gender – specifically the tendency for ‘male’ names to become unisex or even predominantly ‘female’. Today I actually got around to crunching the data!
I found myself with 37,407 names that had been used for males, and 62,318 for females – so we immediately see that there’s a lot more diversity for female names. Of all these, 9800 are common to both genders – but only 8,564 have instances with both genders in the same year. Any such year I could use for a scatter plot of the proportion of males with a given name, out of all people with that name. So a value of 1 indicates a name was assigned only to males that year, with 0 showing that only females received it. Here’s the ratios for `Hilary’:
Proportion of male Hilarys.
This is a little misleading as it suggests that there was always a 10-20% proportion of Hilarys who were female, but pay attention to the scale – there are no years before 1910 that featured both male and female Hilarys. A quick glance at the raw figures shows that there was a lead-in period where only men (orange) might have the name Hilary, before it started being used for both genders, then became predominantly a female (green) name:
Male (orange) and Female (green) Hilarys.
We can also see that Hilary became far more popular as a female name than it ever was for men, before its spectacular crash in the 1990s. This event was even more disastrous for the name’s masculine use – there are no instances at all after 1993 (although I believe the SS data suppresses any name used only a handful of times in a given year, in interests of anonymity).
I tried to find the most significant example of a traditionally male name transitioning to a female one: I measured this by comparing the difference in maximum and minimum ratios through time. Under this measure, ‘Aubrey’ has most thoroughly shifted gender; in 1949 all but one of the sixty (over 98%) Aubreys were male, but of the 7277 Aubreys named in 2011 there’s a mere 142 males (just under two percent).
Proportion of male Aubreys.
Male (orange) and Female (green) Aubreys.
My next goal was to find unisex names – those where the ratio is reliably around 0.5. In fact, there are some names which are perfectly unisex in the sense that every year they’re recorded, they’re found equally as male and female names. But that’s because they’re simply not found that often: no name had more than five years in which both genders used it in this way. So I allowed some wriggle room – an average ratio in the range 0.48 to 0.52, but with the requirement that were at least 50 years in which they appeared for both genders. This turned up sixteen names, although of the two with more than a hundred records, one is ‘Unknown’. The other, though, is ‘Dee’:
Proportion of male Dees.
Male (orange) and Female (green) Dees.
This one’s interesting because the average is more on account of balanced periods of predominantly-male or -female use, rather than balanced use each year. Another from this collection is ‘Paris’- on the basis of 79 shared years, it averages 51.3% male use. But again looking at the plot we see that it favoured just one gender, then shifted to the other:
Proportion of male Paris’
Whenever I’ve mentioned this topic online, someone has cheerfully asserted that this process only works one way – that it’s never the case that a traditionally female name becomes adopted for use with males. The data doesn’t seem to back that up as an absolute- although I’d have to do some more sophisticated statistics to determine trendlines, there are some examples which are immediately obvious. To find them, I looked for names where the difference between maximum and minimum usage as a male name was at least 0.8, and the maxium occurred after the minimum. This recovered 16 names: Alpha, Artie, Audie, Caron, Donnie, Frankie, Gerry, Jan, Jean, Joell, Jullian, Karon, Lashawn, Lavon, Robbie, and Taran. (Without the stipulation of an 80% shift, I find many, many more but they can often be explained by the problem of few common years, rather than a genuine trend.)
For instance, here’s the plot for ‘Donnie’:
Proportion of male Donnies.
Male (orange) and Female (green) Donnies.
Attempting to guess some unisex names revealed cases where a name was strongly male, then was taken over by females, before rallying somewhat (which probably explained why I thought of them as being for either gender):
Proportion of male Kims.
Proportion of male Tracys.
Finally, I wanted to test the other idea I commented on – that British surnames are slowly becoming fair game as American female first names! Using this list of the hundred most common British surnames, I found that all but four were registered as first names at some point. However, all 96 feature as male names, but only 60 as female names – so it seems that I’m wrong – it’s males who are receiving ‘surnames’ as first names originally, and then some of those make the cut as female names. This is all rather confusing for me when I’m in the US, where either of my names works as a first or last name, and my last name works as a first name of either gender! In closing, here are the plots for that:
Proportion of male Taylors.
Male (orange) and Female (green) Taylors.